This project is designed to provide information on basic neural mechanisms involved in the generation and control of respiratory movements in mammals. The long-range goal is to explain the ontogeny and neurogenesis of respiratory movements in terms of the molecular, biophysical, synaptic, and network properties of respiratory neurons in the mammalian brainstem and spinal cord. Current work focuses on cellular and network mechanisms generating the respiratory rhythm in the brainstem. A set of interrelated, multidisciplinary studies are ongoing to determine: sites, cellular components, and architecture of brainstem networks involved in generation and transmission of respiratory rhythm; biophysical properties and synaptic interactions of neurons forming the respiratory oscillator; and neurochemical mechanisms for modulation and synaptic transmission of rhythm. Experiments are performed with isolated in vitro brainstem-spinal cord and brainstem slice preparations from fetal, neonatal, and juvenile rodents. The critical brainstem locus containing the populations of neurons generating the rhythm has been identified. Novel methods were developed to isolate this locus (called the pre-Botzinger complex) in thin slices of the medulla which retain functionally active respiratory networks, allowing experimental analysis of mechanisms concurrently at cellular and network levels. We have further developed novel methods for real-time structural and functional imaging of the rhythm generating neurons utilizing infrared illumination and differential interference contrast (IR-DIC) optics simultaneously with fluorescence imaging of the neurons labeled with calcium-sensitive dyes. This imaging approach has facilitated identification of the rhythm generating neurons for electrophysiological studies of biophysical and synaptic properties. With these approaches, we have imaged the activity of respiratory pacemaker neurons in the pre-Botzinger complex in vitro, providing the most direct experimental evidence to date that rhythm generation involves neurons with specialized pacemaker properties. These results continue to support our hybrid pacemaker-network model which was formulated from previous work to explain rhythm generation. Computational approaches have been used in parallel to experimental studies to model respiratory neurons and networks. Our biophysically realistic computational models of the pacemaker neurons have been refined and novel investigations were conducted on the dynamic behavior of synaptically-coupled populations of these cells. Simulations with these models mimic many features of the neuron activity and synaptic interactions found experimentally in vitro. Computer-based methods have also been further refined to produce animations of these simulations, allowing visualization of the dynamic behavior of the model neurons and their network interactions. These models are currently being applied to further explore and visualize principles of operation of the respiratory oscillator at different stages of nervous system development.